build(deps): bump agents from 0.11.9 to 0.19.0

[dependabot]

Bumps agents from 0.11.9 to 0.19.0.

Release notes

Sourced from agents's releases.

agents@0.19.0

Minor Changes

• #1922 cb4c1c7 Thanks @​cjol! - Support both AI SDK v6 and v7.

  The ai peer range is ai@^6 || ^7 (and @ai-sdk/react is @^3 || ^4) across agents, @cloudflare/ai-chat, @cloudflare/codemode, and @cloudflare/think. Consumers can adopt AI SDK v7 or stay on v6 — no forced AI SDK upgrade when bumping these packages.

  Only the current implementations are covered. New optimisations and APIs made available by AI SDK v7 (broader result-shape audits, stream helper migrations, etc.) are intentionally out of scope.

  How dual-version support works — @cloudflare/think calls the AI SDK through the option names present in both majors (v7 keeps the v6 names as aliases), and normalizes the genuine divergences at the boundary:

  • Uses stepCountIs, system, experimental_telemetry, onStepFinish, and experimental_onToolCallFinish (in v7 this alias resolves to onToolExecutionEnd and fires once).

  • The tool-execution-finished event is normalized across majors: v6's { success, output, error, durationMs, stepNumber } and v7's { toolOutput, toolExecutionMs } collapse to one ToolCallResultContext. stepNumber is undefined under v7 (the v7 event no longer provides it).

  • The UI message stream is built via the result's toUIMessageStream() method (present in both majors); the standalone toUIMessageStream({ stream }) helper and result.stream are v7-only.

  • The workspace read tool emits { type: "file-data", data, mediaType, filename } model output, accepted by both majors (v7's newer { type: "file", data: { type: "data", data } } shape does not exist in v6).

    Public API notes:

  • @cloudflare/think keeps system, onStepFinish, and experimental_telemetry where callers already use them, and also accepts TurnConfig.telemetry (forwarded ahead of experimental_telemetry when present).

  • @cloudflare/ai-chat updates the AIChatAgent.onChatMessage callback type from StreamTextOnFinishCallback to GenerateTextOnFinishCallback.

    Verified against both ai@6 and ai@7: @cloudflare/think type-checks with zero errors and its full workers test suite passes under each major.

    Known limitations:

  • workers-ai-provider (Think's default model provider) is fixed at v4 for AI

... (truncated)

Changelog

Sourced from agents's changelog.

0.19.0

Minor Changes

• #1922 cb4c1c7 Thanks @​cjol! - Support both AI SDK v6 and v7.

  The ai peer range is ai@^6 || ^7 (and @ai-sdk/react is @^3 || ^4) across agents, @cloudflare/ai-chat, @cloudflare/codemode, and @cloudflare/think. Consumers can adopt AI SDK v7 or stay on v6 — no forced AI SDK upgrade when bumping these packages.

  Only the current implementations are covered. New optimisations and APIs made available by AI SDK v7 (broader result-shape audits, stream helper migrations, etc.) are intentionally out of scope.

  How dual-version support works — @cloudflare/think calls the AI SDK through the option names present in both majors (v7 keeps the v6 names as aliases), and normalizes the genuine divergences at the boundary:

  • Uses stepCountIs, system, experimental_telemetry, onStepFinish, and experimental_onToolCallFinish (in v7 this alias resolves to onToolExecutionEnd and fires once).
  • The tool-execution-finished event is normalized across majors: v6's { success, output, error, durationMs, stepNumber } and v7's { toolOutput, toolExecutionMs } collapse to one ToolCallResultContext. stepNumber is undefined under v7 (the v7 event no longer provides it).
  • The UI message stream is built via the result's toUIMessageStream() method (present in both majors); the standalone toUIMessageStream({ stream }) helper and result.stream are v7-only.
  • The workspace read tool emits { type: "file-data", data, mediaType, filename } model output, accepted by both majors (v7's newer { type: "file", data: { type: "data", data } } shape does not exist in v6).

  Public API notes:

  • @cloudflare/think keeps system, onStepFinish, and experimental_telemetry where callers already use them, and also accepts TurnConfig.telemetry (forwarded ahead of experimental_telemetry when present).
  • @cloudflare/ai-chat updates the AIChatAgent.onChatMessage callback type from StreamTextOnFinishCallback to GenerateTextOnFinishCallback.

  Verified against both ai@6 and ai@7: @cloudflare/think type-checks with zero errors and its full workers test suite passes under each major.

  Known limitations:

  • workers-ai-provider (Think's default model provider) is fixed at v4 for AI SDK v7. Consumers on ai@6 who rely on Think's built-in default model may hit a provider-version mismatch; passing their own LanguageModel avoids this.
  • chat@4.31.0 currently declares an ai@^6 peer and does not yet advertise

... (truncated)

Commits

• 4a3ee29 Version Packages (#1976)
• cb4c1c7 feat: update AI SDK to v7 (#1922)
• 83e093b Version Packages (#1943)
• 33e59c4 fix(mcp): recover expired HTTP sessions (#1923)
• c19d58a fix(mcp): harden connection recovery (#1924)
• 6861933 fix(mcp): harden OAuth callback recovery (#1926)
• 8ad5f39 docs(agents): document current ctx.id.name availability and pin it with a col...
• 3ce98ff fix(chat): reconcile errored reconnect streams (#1963)
• f5b1dd8 feat: Cloudflare-native AI tracing (agents/observability/ai) (#1860)
• a3cbed1 fix(mcp): avoid repeated tool schema materialization (#1959)
• Additional commits viewable in compare view